Abstract

Dissolved organic matter (DOM) in freshwater ecosystems has its origin in a multitude of terrestrial and aquatic sources which determine amount, composition and thereby its functions. Spectroscopic methods are used to characterize chromophoric DOM. Among these methods, absorbance and fluorescence spectroscopy can provide indications of sources, behavior, and biogeochemical cycling of DOM. Since DOM is defined as the part of the organic matter pool which passes filters from 0.22 to 0.7 mu m, sample filtration is required before spectroscopic measurements can be done. However, the use of different filter types might be influencing the results of DOM composition measurements. In order to assess the effect of different filters, we used three filter types with nominal pore sizes of 0.7, 0.45 or 0.221 mu m, measured difference spectra, calculated spectroscopic indexes (S275-295, S350-400, S-r, SUVA(254), E2: E3, fluorescence index, beta:alpha, humification index) and fluorescence components (parallel factor analysis, PARAFAC), obtained via spectroscopic measurements from water with different characteristics (e.g. river water, wetland, sewage effluent). Our results show that the filter types had significant but small effects on the chromophoric DOM composition and that meta-analyses of DOM composition based on studies with different pore size can be done, when keeping potential minor filtration effects in mind. This is important for further meta-analysis which might inter-compare data sets on spectroscopic characterization of DOM. (C) 2014 Elsevier GmbH. All rights reserved.